Mining drug-disease relationships:a recommendation system
10.3969/j.issn.1001-1978.2015.12.028
- VernacularTitle:药物-疾病关系预测:一种推荐系统模型
- Author:
Hao WANG
;
Haiping WANG
;
Xindong WU
;
Qi LIU
- Publication Type:Journal Article
- Keywords:
drug repositioning;
biomedical big data;
recommen-dation system;
similarity measures;
collaborative filtering;
drug-disease relationships prediction;
machine learning
- From:
Chinese Pharmacological Bulletin
2015;(12):1770-1774
- CountryChina
- Language:Chinese
-
Abstract:
Aim Drug repositioning is to find new indications for existing drugs,however,potential drug-disease relationships are often hidden in millions of unknown relationship.With the analyzing of medical big data,we predict the potential drug-dis-ease relationships.Methods Based on the assumption that similar drugs tend to have similar indications,we applied a rec-ommendation-based strategy to drug repositioning.First,we col-lected the information of known drug-disease therapeutic effect, side effect,drug characters and disease characters;second,we calculated the drug-drug similarity measurements and disease-disease similarity measurements;last,we used a collaborative filtering (CF)method to predict unknown drug-disease relation-ships based on the known drug-disease relationships by integra-ting the similarity measurements,and built a ranking list of pre-diction results.Results The experiments demonstrated that a-mong the TOP 500 of the list,1 2.8% were supported by clinical experiments or review,and 20% were supported by model or-ganism or cell experiments.Conclusion Compared to the clas-sification model and random sampling results,the CF model can effectively reduce the false positives.